ClawHub
Back to skill

Security audit

tongban government policy understanding

Security checks across malware telemetry and agentic risk

Overview

This skill is a public Shanghai government-service lookup guide with disclosed local references and optional public-data refresh scripts, but users should verify important official-process advice.

Install only if you want a directional Shanghai government-service guide. Verify official URLs, 随申办, or 021-12345 before acting, especially for licensing, medical, deadline-sensitive, or eligibility questions. Run the refresh scripts only deliberately because they fetch public data and rewrite local reference files.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (2)

Lp3

Medium
Category
MCP Least Privilege
Confidence
86% confidence
Finding
The skill declares no permissions, yet its instructions explicitly direct use of local file reads (`references/*`), network access to the official site, and scripts that write refreshed data to disk. This mismatch is a real security/control issue because the runtime capabilities exceed what is transparently declared, making review, sandboxing, and least-privilege enforcement harder.

Intent-Code Divergence

Medium
Confidence
95% confidence
Finding
The entry maps the broad user query "医师注册" to the much narrower service "人体器官移植执业医师注册", which can misroute users to the wrong government procedure. In a public-service guidance skill, incorrect reference mapping can cause users to rely on inaccurate official-process advice, miss required steps, or submit to the wrong department.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Static analysis

No suspicious patterns detected.